Provided by the author(s) and University College Dublin Library in accordance with publisher policies. Please
cite the published version when available.
Title
Author(s)
Characteristics and mechanisms of phosphate adsorption on
dewatered alum sludge
Yang, Y.; Zhao, Y.Q.; Babatunde, A.O.; Wang, L.; Ren, Y.X.;
Han, Y.
Publication
date
2006-09
Publication
information
Separation and Purification Technology, 51 (2): 193-200
Publisher
Elsevier
Link to online http://dx.doi.org/10.1016/j.seppur.2006.01.013
version
Item
record/more
information
http://hdl.handle.net/10197/3177
Publisher's
statement
This is the author’s version of a work that was accepted for
publication in Separation and Purification Technology.
Changes resulting from the publishing process, such as peer
review, editing, corrections, structural formatting, and other
quality control mechanisms may not be reflected in this
document. Changes may have been made to this work since it
was submitted for publication. A definitive version was
subsequently published in Separation and Purification
Technology, 51 (2): 193-200 DOI
10.1016/j.seppur.2006.01.013.
Publisher's
version (DOI) http://dx.doi.org/10.1016/j.seppur.2006.01.013
Downloaded 2017-06-16T08:20:52Z
The UCD community has made this article openly available. Please share how this access benefits you. Your
story matters! (@ucd_oa)
Some rights reserved. For more information, please see the item record link above.
Characteristics and mechanisms
of phosphate adsorption on dewatered alum sludge
Y. Yanga, Y. Q. Zhaoa, *, A. O. Babatundea, L. Wangb, Y. X. Renb, Y. Hanb
a
Centre for Water Resources Research, School of Architecture, Landscape and Civil
Engineering, University College Dublin, Earlsfort Terrace, Dublin 2, Ireland
b
*
School of Environmental and Municipal Engineering, Xi’an University of
Architecture and Technology, Xi’an, 710055, P. R. China
Corresponding author:
Y. Q. Zhao
Centre for Water Resources Research, School of Architecture, Landscape and Civil
Engineering, University College Dublin, Earlsfort Terrace, Dublin 2, Ireland
Tel:+353-1-7167284
Fax:+353-1-7167399
E-mail: [email protected]
1
Abstract
The adsorption characteristics of phosphate adsorption on the dewatered alum
sludge were identified as a function of pH and ion strengths in solution. In addition,
adsorption mechanisms were investigated by conducting batch tests on both the
hydrolysis and P-adsorption process of the alum sludge, and making a comparative
analysis to gain newer insights into understanding the adsorption process. Results
show that the adsorption capacity decreased from 3.5 to 0.7 mg-P/g-sludge when the
solution pH was increased from 4.3 to 9.0, indicating that adsorption capacity is
largely dependent upon the pH of the system. The results of the competitive
adsorption between phosphate and typical anions found in wastewater, such as SO42and Cl-, onto alum sludge reveal that alum sludge can selectively adsorb phosphate
ions. The insignificant effect of SO42- and Cl- on P-adsorption capacity indicates that
phosphate adsorption is through a kind of inner-sphere complex reaction. During the
adsorption process, the decrease of phosphate concentration in solution accompanied
with an increase in pH values and concentrations of SO42-, Cl- and TOC (total organic
carbon) suggests that phosphate replaced the functional groups from the surface of
alum sludge which infers that ligand exchange is the dominating mechanism for
phosphate removal. At the same time, the simultaneous decreases in PO43- and total
aluminium concentration in solution indicate that chemical reaction and precipitation
are other mechanisms of phosphate removal.
Keywords: Adsorption mechanisms; Alum sludge; Ligand exchange; Phosphate
removal; Surface complex
2
1. Introduction
Alum sludge is an inescapable by-product of the processing of drinking water in
water treatment works where aluminium salt is used as the coagulant. Aluminium
sulphate is arguably the most widely used coagulant in drinking water treatment.
When aluminium sulphate is added to raw water, it dissociates into Al3+ and SO42-.
The
Al3+
ions
are
immediately
surrounded
by
water
molecules
and
hexaaquoaluminium ([Al(H2O)6]3+) is rapidly formed. The hexaaquoaluminium
formed then undergoes series of rapid hydrolytic reactions to form charged polymeric
or oligomeric hydroxo-complexes of various structures. Such hydrolytic products
include [Al(H2O)5OH]2+, [Al(H2O)4(OH)2]+, [Al6(OH)15]3+, [Al8(OH)20]4+ and
Al(OH)3(s) etc. [1]. During coagulation in water treatment process, these complexes
adsorb and modify the surface charge of the colloidal particles, e.g. natural organic
matter (NOM) such as humic and fulvic acid, microorganisms such as bacteria,
protozoa and algae, and inorganic substances such as fine soil particles [2]. Thereafter,
in the ensuing treatment units including flocculation, sedimentation and filtration, the
colloids in the raw water are removed and transferred to the sludge phase together
with the hydrolytic aluminium species.
In recent years, the management of alum sludge has become a significant issue in
environmental engineering due to the enormous quantities generated and the
associated disposal costs and constraints. In Ireland alone, a double fold increase in
alum sludge generation has been forecast by the end of the next decade, from a
current estimate of 15,000 to 18,000 t/pa of the dewatered solids. However, as a
sustainable approach to mitigate these effects, current trends have indicated a
progressive drive towards alum sludge reuse as beneficial material. Such beneficial
reuses include the use of alum sludge as an adsorbent for phosphorus removal from
wastewaters. In particular, previous works done by the authors showed that the Irish
dewatered alum sludge has a latent adsorption capacity, and it can be utilized as a
“low-cost” phosphorus sorption medium in wastewater treatment [3,4]. In other
instances, conjunctive evidences from literature have also shown that alum sludge can
help remove phosphorus in wastewater. This is attributed to the abundant aluminum
3
ions in the alum sludge, which enhance the processes of adsorption and chemical
precipitation that help to remove such pollutants from wastewater [5-7]. However, an
in-depth understanding of the mechanisms and characteristics of phosphorus
adsorption by the alum sludge is crucial to its effective utilization as an adsorbent
material.
Therefore, the aims of this study are (1) to investigate the phosphate adsorption
mechanisms of a dewatered alum sludge, (2) to identify the factors that affect the
phosphate adsorption capacity of the alum sludge and (3) to examine the structural
identity of the phosphate impregnated alum sludge (PIAS).
2. Materials and methods
2.1. Preparation of test materials
Dewatered alum sludge cake (moisture content 72~75%) was collected from an
industrial filter press of the sludge dewatering unit of a Water Treatment Works in
Southwest Dublin, Ireland where aluminium sulphate is used as coagulant. After
collection, the sludge cakes were air-dried and the moisture content decreased to
10.2% at the time of being used. The air-dried sludge was then ground and sieved to
provide the test adsorbent with diameter < 0.063mm.
Artificial wastewater was synthesized by dissolving pre-weighed potassium
dihydrogen phosphate (KH2PO4) in distilled water. The solution was then incubated in
the laboratory at 20±2 oC, and adjusted to different pH (using sulphuric acid (0.01M)
and sodium hydroxide (0.1M)). Solutions were kept airtight to prevent CO2 from
affecting solution pH.
Chloride stock solution (6.0M) and sulphate stock solution (3.0M) were prepared
by dissolving pre-weighed amount of NaCl and Na2SO4 respectively, in distilled
water.
2.2. Characterization of the alum sludge
Elemental, physical and chemical analyses of the dewatered alum sludge were
carried out using Inductively Coupled Plasma-Atomic Emission Spectrometer
4
(ICP-AES,
IRIS
Advantage),
TOC-V
CSH
(Shimadzu)
and
Fourier
Transform-infrared (FT-IR, EQUINOX-55). The morphological structure of the
dewatered alum sludge was examined by X-ray diffractometer (XRD, D/max-3C) and
Scanning Electron Microscope (SEM, JSM-6700F).
2.3. Adsorption characteristics tests
2.3.1. Adsorption capacity
Batch tests were conducted by adding pre-determined amount of the sludge to a
pH controlled distilled water (pH 4.3 to 9) and measuring the zeta potential. Sulphuric
acid (0.01 M) and sodium hydroxide (0.1M) were used to control the pH of the
adsorption system to designed values (pH 4.3 to 9) to investigate the effect of pH on
adsorption behaviour/capacity and the amount of acid/alkaline required to maintain
the pH was calculated from the concentration and volume added, including a
correction for the dilution effect. Thereafter, standard aliquots of the phosphate stock
solution was introduced, giving a resultant phosphate concentration of 5 mg-P/l,
which simulated the P level in typical municipal wastewater [8]. The mixtures were
then mechanically agitated for promoting adsorption over a 48 hour pre-determined
equilibrium time [3]. After adsorption, equilibrated samples were filtered using 0.45
µm millipore filter paper (Whatman) and analysed for phosphate concentration.
Adsorption capacities were evaluated from a linearized form of the Langmuir
adsorption isotherm
Ce
1
1

Ce 
q Q0
Q0 b
(1)
where q is the mass of phosphorus adsorbed per unit mass of sludge; Ce is the
equilibrium concentration of phosphorus (mg-P/l) in suspension after adsorption; Q0
is the maximum adsorption capacity (mg-P/g-sludge); and b is a constant related to
the energy of the adsorption-desorption process with unit of inverse of concentration
Ce. By plotting Ce vs. Ce /q, a straight line with slope 1/Q0 is obtained and the Q0 can
5
be calculated.
Phosphate concentrations and pH values were obtained using a Hach
spectrophotometer (DR/2400) and a pH meter (WTW, pH 325) respectively. The Zeta
potential of the alum sludge was measured at different pH using a Zeta Potential
Analyser (ZC-2000). All the adsorption tests were repeated twice and the average
value of measurements was reported.
2.3.2. Anion selectivity of alum sludge and structural identity of PIAS
In order to investigate the selective adsorption of Cl-, SO42- and PO43- by the alum
sludge, pre-determined amount of the sludge was added to a pH modified distilled
water (pH 4.3 to 9) followed by the introduction of Cl- and SO42- at concentrations
ranging from 300 to 1000 mg/l. The total volume of the mixture was noted. Thereafter,
10 mg-P/l of equal volume as the initial mixture volume was added giving a resultant
P concentration of 5 mg-P/l. The final mixture was then mechanically agitated for 48
hours equilibrium time and residual phosphorus concentration determined as in
section 2.3.1. The structural identity of the PIAS was studied by investigating the
effect of the ionic strength on the equilibrium of the surface complex formed.
2.4. Adsorption mechanism
2.4.1. Ligand exchange
To examine the possibility of ligand exchange in the adsorption mechanism,
samples of alum sludge (5g/l) were hydrolyzed in distilled water (initial pH 7.0±0.02)
in an airtight container and left for 28 days (final pH 5.98). Thereafter, phosphate
concentration in the hydrolysed suspended solution was brought up to 5mg-P/l and
phosphate adsorption by the hydrolyzed alum sludge was then monitored for 6 hours.
Filtered samples from both processes were analyzed for Cl-, SO42- and PO43concentration with the use of ICP-AES and Ion Chromatography (DX-120) while the
pH of the suspended solution was also measured in both cases. Insights into the
mechanisms of phosphate adsorption can be gained by measurement of the amount of
OH- and other anions/substances released from alum sludge consequent upon the
6
adsorption of phosphate.
2.4.2. Interaction between dissolved aluminium species and PO43To investigate interaction between dissolved aluminium species and phosphate
ions, the concentration of total aluminium (including Al3+, Al(OH)2+, Al(OH)2+,
Al(OH)3 and Al(OH)4- etc.), before and after phosphate addition (as in test 2.4.1) was
measured. Particularly, any change in concentration may possibly give an indication
of a reaction taking place. The concentration of total aluminum was accurately
determined with the use of ICP-AES and Ion Chromatography.
2.4.3. Competition with humic substances
The concentration of TOC, before and after phosphate addition (as in test 2.4.1)
was measured in order to understand likely competition between phosphate ions and
humic substances on the surface of alum sludge. Again, a change in the TOC
concentration may evidently support the competition hypothesis.
3. Results
3.1. Properties of alum sludge
3.1.1. Chemical composition
The principal chemical compositions of the dewatered alum sludge are shown in
Table 1. Although the properties of such sludge are highly variable and dependent on
both the type of the raw water and the chemical composition of coagulant [9], it can
be seen from Table 1 that aluminum is the dominant component in the dewatered alum
sludge with ~46% in mass expressed as Al2O3. The other principal chemical
components were Fe3+, Ca2+, Mg2+, Cl-, SO42- and SiO42-, with mass percentages of
1.7% for cationic ions and 3.6% for anionic ions. In addition, the FT-IR has identified
that the alum sludge contains some organic matter and the content expressed in TOC
was 9.9% by mass examined by a Shimadzu TOC-V CSH total organic carbon
analyzer.
7
[Table 1. The principal chemical compositions of alum sludge]
3.1.2. Morphological structure
XRD was used to identify the morphological structure of the alum sludge
although it cannot provide a quantitative description of such structure [10]. The result
of XRD pattern shown in Fig. 1 illustrated no sharpening characteristic diffraction
peaks over a broad range of d-spacings (15-70º 2θ), indicating poorly ordered
particles within alum sludge. In addition, scanning electron microscope (SEM)
observation of powdered alum sludge, shown in Fig. 2, has not found the classical
well-crystalline appearance on sludge surface. Although both the XRD and SEM are
rather qualitative descriptions on sludge structure, considering Fig. 1 and 2 it is
reasonable to believe that the alum sludge is really amorphous.
[Fig. 1. X-ray diffraction pattern of dewatered alum sludge]
[Fig. 2. SEM observation of dewatered alum sludge]
3.2. Adsorption characteristics
3.2.1. Adsorption capacity
The experimental data from the adsorption tests fit well the Langmuir behaviour,
as illustrated in Fig. 3. Computed results of adsorption capacities at different pH
together with the results of Zeta potential measurement are shown in Fig. 4. It can be
seen that pH has a significant effect on the adsorption capacity. A pH increase from
4.3 to 9.0 remarkably decreased the P-adsorption capacity from 3.5 to
0.7mg-P/g-sludge. The measured zeta potential shows the variation of surface charge
from +75.8 mV to –33.7 mV, corresponding to a pH change from 4.3 to 9.0.
[Fig. 3. Langmuir isotherm plots to determine the maximum adsorption capacities]
8
[Fig. 4. Effect of pH on adsorption capacity and surface charge of alum sludge]
3.2.2. Selective adsorptivity of alum sludge and structural identity of PIAS
The effects of Cl- and SO42- on the phosphate adsorption capacity of the alum
sludge are shown in Fig. 5. At pH 4.3, despite the wide variation in the concentrations
of Cl- and SO42- from 300 to 1000 mg/l, there was no significant difference in the
adsorption capacity of the alum sludge which ranged from 3.5 to 3.4 mg-P/g-sludge.
The same insignificant difference was observed at all the other pH values. This result
is a good agreement with the findings reported by Tanada et al. [11] in which the
selectivity of phosphate adsorption onto pure aluminium oxide hydroxide was
evaluated to be about 7000 and 260 times compared with that of Cl- and SO42-,
respectively. Fig. 6 illustrates the FT-IR spectra of the sludge samples before and
after adsorption for the purpose to explore the identity of PIAS. Unfortunately, by
examining the spectral region from 1200-800cm-1, which was proposed to be the
region for structural diagnosis of metal orthophosphate complexes in aqueous
solutions [12], a nearly identical spectrum can be observed from Fig. 6. The reason of
not identifying phosphate on the PIAS by FT-IR is unclear, but having something
linked with the complication of phosphate-alum adsorption behaviour in alum sludge.
Unlike the phosphate-alum adsorption on pure aluminium oxide hydroxide, alum
sludge contains some other ions and organic matter, which may interfere the FT-IP
results.
[Fig. 5. Effects of Cl- and SO42- on phosphate adsorption capacity at different pH]
[Fig. 6. The FT-IR spectra of the sludge samples before and after adsorption]
3.3. Adsorption mechanisms
Results of the pre-hydrolysis process of the alum sludge followed by the
9
P-adsorption process are shown in Fig. 7. Emphasis was on the changes in the pH and
concentrations of Cl- SO42-, TOC, and total aluminium during the hydrolysis process
of the alum sludge and the immediate P-adsorption process. The results show that
alum sludge exhibited a strong hydrolysis potential characterized by an initial rapid
release of H+, Cl-, SO42-, TOC and total aluminium in the first 24 hours followed by a
slow release, which was monitored until an equilibrium concentration of each
ion/substance was reached. Thereafter phosphate was added on 28th day to initialize
the adsorption process.
During the hydrolysis stage, a decrease in pH from 7.0 to 5.98 was observed
indicating complex reactions between the alum sludge and OH- ion, resulting in a
release of H+ ions, which leads to the decrease in pH. Concurrently, the concentrations
of Cl-, SO42-, TOC and total aluminum increased from 0 to 3.13 mg/l, 2.45 mg/l, 5.38
mg/l and 33µg/l, respectively, as the hydrolysis progressed (Fig. 7, hydrolysis stage).
However, during the adsorption stage, the reduction in phosphate concentration
occurred simultaneously with increases in pH (5.98 to 7.21), and concentrations of Cl(3.13 to 7.89mg/l), SO42- (2.45 to 3.67 mg/l) and TOC (5.38 to 9.16 mg/l) coupled
with a decrease in the concentration of total aluminium from 33 to 23 µg/l (Fig. 7,
adsorption stage). These results give valuable insights that are vital to the
understanding of the adsorption mechanism and which are further discussed in this
paper.
[Fig. 7. Variation of pH, Cl-, SO42-, TOC and total aluminium during the hydrolysis
and P-adsorption process]
4. Discussion
4.1. Functional groups on the surface of alum sludge
The decrease of pH from 7.0 to 5.98 during the hydrolysis stage of the alum
sludge (Fig. 7, b) shows that alum sludge has a particularly strong tendency to
hydrolyze in water. Result of the Zeta potential measurement of the alum sludge at pH
7.0 gave a value of +28 mV, which indicates the presence of unsatisfied positive
10
charges on the surface of the alum sludge. Drawing on similar findings by Hem and
Roberson [13], when alum sludge is added to water, it can be expected that the alum
sludge is surrounded by a tightly bound shell of oriented water molecules (HO–H) on
its surface and the positive charge on the alum sludge tends to weaken the forces
holding the protons (–H) to the oxygen, and thus the protons are relatively easy to
release. The release of H+ causes the decrease of pH which suggests a change in the
number of OH bonded on the surface of the alum sludge, leading to a hydroxylated
surface [13]. These surface hydroxyl groups can take part in complexation reactions
with metal ions and ligands [14].
At the same time, the decrease in pH was accompanied by an increase in the
concentration of Cl-, SO42- and humic substances (as TOC) in solution (Fig. 7 c & d,
hydrolysis stage). The change in the concentration of these ion/substances is
attributable to an ion exchange process between –OH (in solution) and Cl-, SO42- and
humic substances (on the surface of the alum sludge) to form a new counter-ion layer
on the surface of the alum sludge. The reactions occurring during the hydrolysis stage
can be possibly explained by equations (2, 3, 4 & 5).
Al
HO_H
Al
(
OH + H+
2)
Al
_
Cl + HO H
(
OH + Cl- + H+
Al
3)
(
Al ) 2 SO4 + 2 HO - H
Al
2
OH + SO42- + 2 H+
(
4)
(
Al
_
Humic + HO H
Al
+
OH + Humic + H
5)
It is evident from the hydrolysis stage that there are some functional groups on
the alum sludge surface, such as –OH, –Cl, –SO4 and humic substances, which are a
kind of an activated group, and have the potential to be selectively exchanged with the
alum sludge affinity anion. In addition, if all the increases of Cl- and SO42concentrations were due to equations (2, 3 and 4), then the calculated pH of the
11
solution should decrease from 7.0 to 3.85. However, the change of pH during the
hydrolysis stage was found to be 7.0 to 5.98 (Fig 7 b, hydrolysis stage). This means
that there are still some other mechanisms involved in the increase of Cl- and SO42concentrations, which could be the dissolution of chlorine or sulphate salts during the
hydrolysis stage, such as CaCl2, FeCl3, MgCl2, CaSO4, Fe2(SO4)3 and MgSO4 etc
since the alum sludge contains such cations (see Table 1).
4.2. The pH dependence of adsorption capacity
The results of the effect of pH on adsorption capacity show that the adsorption
capacity of alum sludge is strongly dependent on solution pH and several reasons may
be adduced to this. First, the change of surface characteristics can affect adsorption
capacity to some extent. The surface characteristics of alum sludge, such as the
surface charge (as related to the zeta potential measurements) as shown in Fig. 4
resulted from proton transfers at the surface [15]. It can be seen from Fig. 4 that as pH
increases, the surface charge of the alum sludge, which is pH dependent [16], changes
from positive to negative. This gives valuable information regarding the changing
nature of the surface of the alum sludge. The increase of OH- in solution could affect
the electrostatic properties of the alum sludge. For example, the adsorption of –OH on
the surface of alum sludge leads to the formation of a new charged counter-ion layer,
making the alum sludge surface being a comparable lower affinity to phosphate.
Again, with an increase in pH, OH- competes strongly with phosphate for active sites,
which, in turn affects the adsorption capacity of the alum sludge. Second, the pH
dependence of phosphate adsorption capacity may be associated with ligand exchange
between phosphate ions (in solution) and –OH (on the surface of alum sludge). One
characteristic feature of phosphate adsorption on hydrous aluminium oxide is the
release of hydroxyl ions into the solution [17]. This therefore implies that adsorption
is favored by low pH values and that adsorption capacity would be higher at low pH
values than at high pH values. This agrees with the report of Kim et al. [7] on the
effect of pH on phosphate adsorption, which shows a lower pH being favourable to
12
phosphorus adsorption.
4.3. Structural identity of PIAS
With regards to further adsorption characteristics of phosphate on the surface of
alum sludge, the result of the effect of anions on adsorption capacity is particularly
useful in determining whether the phosphate is associated with the surface of the alum
sludge as an inner-sphere or outer-sphere complex. It is important to distinguish
between outer-sphere and inner-sphere complex in order to understand the different
stability of adsorbed ion, and the various chemical and physicochemical properties of
alum sludge. Generally, direct evidence for inner-sphere complex can be obtained
from spectroscopic methods, such as Fourier transform infrared spectroscopy [18,19].
However, the FT-IR results of PIAS obtained in this study did not show any
distinctive characteristic peaks (see Fig. 6). Therefore, the structural identity of PIAS
cannot be determined by the FT-IR method. However, a simple method of
distinguishing between inner-sphere and outer-sphere complex is to assess the effect
of ionic strength on the surface complex formation equilibrium. A strong dependence
of the surface complex formation equilibrium on ionic strength is typical for an
outer-sphere complex while the inverse is also implied [2, 20]. In this study, the
insignificant effects of Cl- and SO42- at different concentrations on the adsorption
capacity (see Fig. 5) reflect that the ionic strength does not affect the adsorption
behaviour, indicating an inner-sphere complex. Therefore, it is reasonable to speculate
that the phosphate is adsorbed on the surface of alum sludge by chemical bond.
4.4. Adsorption mechanisms
4.4.1. Ligand exchange
Unlike the hydrolysis process, the rapid increase of pH from 5.98 to 7.12 coupled
with the rapid decrease of phosphate concentration in 1 hour (Fig. 7 a & b) in
P-adsorption process. The increase of pH is a sign of the release of hydroxyl ion from
alum sludge into the solution; this indicating that phosphate is adsorbed by surface
complex, i.e. the ligand exchange between phosphate (in solution) and –OH (on alum
13
sludge surface). The OH involved in the ligand exchange process is a kind of
nonstructural OH, formed during both the formation process of alum sludge in
drinking water treatment and the hydrolysis process. According to the calculation on
the basis of chemical equilibrium reaction and constant of the phosphate speciation
[21] in solution at pH 5.98 when the P-adsorption process took place, the
approximately 96.5% of phosphate in solution existed as H2PO4-. Therefore the
adsorption process could be described by equation (6).
2
Al OH+ H2PO4-
(
Al )2HPO4 + H2O + OH-
(
6)
Quantitative relationship between adsorbed phosphate and hydroxyl ions released
from alum sludge (Eq. (6)), would suppose that if the OH ligand exchange were the
sole adsorption mechanism, the release of hydroxyl ion into the solution would
increase the solution pH from 5.98 to at least 10.2. However, the pH only increased
from 5.98 to a maximum of 7.21 (Fig. 7 b, adsorption stage). This limited increase of
pH implies that there are some other mechanisms contributing to the phosphate
adsorption. The fact that the decrease of phosphate accompanied with the increase of
Cl- and SO42- in solution, as shown in Fig. 7 c, indicates that the alternative
mechanism of P-adsorption by alum sludge could be the ligand exchange between
phosphate (in solution) and Cl- and SO42- (on alum sludge shown in Table 1).
P-adsorption based on this mechanism accompanied with the release of H+ to the
solution could cause a limitation on pH increase. This proposed ligand exchange
process can be shown as equations (7 & 8).
2
Al Cl + H2PO4-
(
Al ) 2 SO4 + H2PO4-
(
Al )2HPO4 + 2 Cl- + H+
(
Al )2HPO4 + SO42- + H+
(7)
(8)
Although the ligand exchange may be one of the major mechanisms of
14
phosphate-alum adsorption, it should be pointed out that the P-adsorption is a
complicated process due to the differences in the formation of phosphate-alum
complexes. Rajan [17] and Guan et al. [22] concluded that phosphate adsorbs onto
metal hydroxide by forming not only monodentate complexes, but also bidentate or
binuclear complexes. Different P species may have different adsorption behaviours on
the same adsorbent. For ortho-P source used in this study, Guan et al. [23] claimed
that the ortho-P tended to form monodentate complex with aluminium hydroxide
surface in adsorption process. Thus, the P-adsorption with Al in a real adsorption
system could be far more complicated. In addition, it is noted in Fig. 7 b that a slow
increase of pH after 1 hour adsorption was observed. This is a sign of the slow release
of hydroxyl ion, leading to a slow decrease of P in solution (see Fig. 7 a), implying the
slow P-adsorption process. Makris et al. [24] opined that the slow P-adsorption
process of drinking water treatment plant sludge might be explained by intraparticle
phosphate diffusion in micropores.
4.4.2. Competitive effect with humic substances
Based on the fact that the increase of humic substances (TOC) was positively
correlated with the removal of phosphate in solution (Fig. 7 d, adsorption stage), the
release of humic substances to the solution would be due to the competitive effect of
phosphate for surface sites with the humic substances. Therefore, as a result, part of
the functional humic substances on the alum sludge surface is replaced by the
phosphate, as shown in equation (9). Similar findings were reported in studies on
competition between phosphate ion and humic substances for active sites during
adsorption process [25]. In spite of the release of TOC in both the hydrolysis and the
adsorption processes, it is noted that the increase of TOC in the solution was from
5.38 to 9.16 mg/l during the P-adsorption stage. The released TOC in the tested pH
range (5.98-7.21) compared with the TOC in alum sludge (see Table 1), represents
just 3.9% of the TOC contained in the sludge. In other words, the release of TOC from
alum sludge is not a significant issue as far as effluent quality is concerned,
15
particularly when alum sludge is used as an adsorbent for phosphate removal in
practice.
2
Al
Humic+ H2PO4-
(
Al )2HPO4 + 2 Humic + H+
(
9)
4.4.3. Chemical reaction and coprecipitation
The variation of soluble aluminium hydrolysis products (as total aluminium)
during sludge hydrolysis and adsorption processes shown in Fig. 7 e, reveals that: (1)
the magnitude of aluminium release from hydrolysis is insignificant although
aluminium is the major component in the alum sludge (Table 1). This implies that the
aluminium in the alum sludge is in a stable and immobilized form, at the test pH value
(5.98-7.21). It should be pointed out that the calculated solubility limit of the “total
aluminium” by considering aluminium phosphate and aluminium hydroxide species is
56.4 µg/l. However, monitored "total aluminium" is 23µg/l as shown in Fig. 7 e. The
difference could be attributed to the multi-ions interactions in the system; (2)
chemical reactions and coprecipitation occur between phosphate and aqueous
aluminium since a decrease in total aluminium concentration from 33 to 23µg/l is
accompanied by a decrease of phosphate concentration. The precipitation formed in
this process is thermodynamically and kinetically favored over aluminium hydroxide
precipitation [26]. As pointed out by Hsu [27] and Omoike and Vanloon [8], the
precipitation is governed by the integration of Al–OH–Al and Al–PO4–Al types of
linkages into an aluminium hydroxyphosphate complex, rather than by precipitation
of discrete phases such as Al(OH)3 or AlPO4; (3) the dissolution of the alum sludge
result in the increase of total aluminium which could provide a continuous supply of
aluminium for the chemical reaction and the coprecipitation process. The intricacies
of the phosphate-alum sludge adsorption system make a distinctive quantitative
analysis of the amount of adsorption and chemical coprecipitation rather difficult.
16
5. Conclusions
Results obtained from this study have provided valuable information regarding
the characteristics and mechanism of phosphate adsorption by alum sludge. In
particular,
(1) The maximum adsorption capacities ranged from 0.7 to 3.5 mg-P/g when the
pH of the synthetic phosphate solution was varied from 9.0 to 4.3. Solution pH has
been shown and established as a vital factor influencing the adsorption behaviour. It is
believed that phosphate adsorption by alum sludge is highly dependent on solution pH
and the surface characteristics of the alum sludge.
(2) The results from the hydrolysis process of alum sludge provides evidence that
the surface of alum sludge contain a significant amount of reactive functional groups,
such as –OH, –Cl, –SO4 and humic substances. These functional groups are
responsible for the ligand exchange mechanism of the adsorption of phosphate onto
the surface of alum sludge.
(3) Alum sludge has a higher selective affinity to adsorb phosphate than typical
anions found in wastewater, such as Cl- and SO42- and the adsorption of phosphate on
the surface of alum sludge is shown to be a kind of inner-sphere complex (“chemical
bond”).
(4) Ligand-exchange is shown to be the dominating mechanism based on
exploratory evidence from the adsorption mechanism of phosphate onto the alum
sludge. Although chemical reaction between phosphate and dissolved aluminium was
demonstrated, it is believed that the chemical reaction played only a marginal role in
the phosphate removal process.
(5) Increased phosphate adsorption coupled with a release of humic substances
indicates the desorption of organic matter as a result of the competition between
phosphate and humic substances for surface site. Experiments also revealed that the
release of humic substances from the alum sludge is not significant at the tested pH
condition (5.98-7.21), thus eliminating concerns over the practicality of reusing
dewatered alum sludge as an adsorbent.
17
Acknowledgements
The authors gratefully acknowledge financial support received from the EPA
(Ireland) Environmental Technologies Scheme (Project NO.: 2005-ET-S-&-M3) and
technical support from the University College Dublin (UCD), Ireland and Xi’an
University of Architecture and Technology, P. R. China. Mr. Patrick Kearney, Section
head technician, Water and Effluents Laboratory, UCD, is thanked for his invaluable
technical assistance during this study. Mr D. Tomlinson and S. Kennedy, graduate
students of UCD, are also thanked for their assistance in the experiments. Sincere
thanks are given to the reviewers for their valuable comments to improve the paper.
References
[1] J. Bratby, Coagulation and flocculation, Uplands Press Ltd., Croydon, 1980.
[2] W., Stumm and J. J. Morgan, Aquatic Chemistry (Third edition), A
Wiley-interscience Publication, New York, 1996.
[3] Y. Yang, D. Tomlinson, S. Kennedy, and Y. Q. Zhao, In Proceedings of the
IWA-WISA conference on: Management of residues emanating from water and
wastewater treatment, South Africa, 2005.
[4] A. O. Babatunde, Y. Yang, and Y. Q. Zhao, In Proceedings of the 10th European
Biosolids and biowastes conference, UK, 2005.
[5] M. G. Sujana, R. S. Thakur and S. B. Rao, Journal of Colloid and Interface Science,
206 (1998), 94-101.
[6] S. H. Huang, and B. Chiswell, Wat. Sci. & Tech., 42 (2000), 295-300.
[7] J. G. Kim, J. H. Kim, H. Moon, C. Chon and J. S. Ahn, Chemical Speciation and
Bioavailability, 14 (2003) 67-73.
[8] A. I., Omoike and G. W. Vanloon, Wat. Res., 33 (1999) 3617-3627.
[9] R. B. Sotero-Santos, O. Rochaa, and J. Povinelli, Wat. Res., 39 (2005) 3909-3917.
[10] Jacob Berkowitz, Michael A. Anderson and Robert C. Graham, Wat. Res., 39
(2005) 3918-3928.
[11] Seiki Tanada, Mineaki Kabayama, Naohito Kawasaki, Toru Sakiyama, Takeo
Nakamura, Mamiko Araki and Takamichi Tamura, Colloid and Interface Science, 257
18
(2003), 135-140.
[12] M. Isabel Tejedor-Tejedor and Mare A. Anderson, Langmuir, 6(1990) 602-611.
[13] J. D. Hem, and C. E. Roberson, Chemistry of aluminum in natural water, United
States Government Printing office, Washington, 1967.
[14] E. Laiti, P. Persson and L.-O. Ohman, Langmuir, 12 (1996) 2969-2975.
[15] W. Stumm, Aquatic surface chemistry, A Wiley-interscience Publication, New
York, 1987.
[16] B. P. Singh, R. Menchavez, C. Takai, M. Fuji, and M. Takahashi, Journal of
Colloid and Interface Science, 291(2005) 181-186.
[17] S. S. S. Rajan, Nature, 253 (1975) 434-436.
[18] G. Lefevre, Advances in Colloid and Interface Science, 107 (2004) 109-123.
[19] P. Persson and K. Axe, Geochimica et Cosmochimica Acta, 69 (2005) 541-552.
[20] Jeanine S. Geelhoed, Tjisse Hiemstra and Willem H. Van Riemsdijk, Geochimica
et Cosmochimica Acta, 61 (1997) 2389-2396.
[21] M. M. Benjamin, Water chemistry, McGraw-Hill, Boston, 2002.
[22] Xiao-Hong Guan, Qing Liu, Guang-Hao Chen and Chii Shang, Colloid and
Interface Science, 289 (2005) 319-327.
[23] Xiao-Hong Guan, Guang-Hao Chen and Chii Shang, Soil Science, 170 (2005)
1-9.
[24] K. C. Makris, W. G. Harris, G. A. O'Connor, and T. A. Obreza, Environ. Sci.
Technol., 38 (2004), 6590-6596.
[25] R. Giesler, T. Anderssonb, L. Lovgrenb, and P. Perssonb, Soil Sci. Soc. Am. J.,
69 (2005) 78-86.
[26] D. Jenkins, J. F. Ferguson and A. B. Menar, Wat. Res., 5 (1971), 369-389.
[27] P. H. Hsu, Wat. Res., 9 (1975) 1155-1161.
19
Table 1. The principal chemical compositions of alum sludge
Chemical composition
Amount (mg/g-sludge)
Aluminum (as Al2O3)
458-463
Iron (as Fe2O3)
11.9-12.3
Calcium (as CaO )
11.6-11.7
Magnesium (as MgO)
7.4-7.6
Humic acid (as TOC)
96.4-98.5
Cl-
16.0-16.2
SO42-
8.2-8.4
SiO42-
10.6-11.8
H2O at 105 oC (moisture content)
H2O at 1000 oC
102
260-270
Note: The range of value indicates the lowest to highest values of the 3 parallel
measurements. Other trace elements analyzed were not reported due to the very small
amount.
20
Strength (cps)
400
300
200
100
0
15
30
45
60
75
2θ (°)
Fig. 1. X-ray diffraction pattern of dewatered alum sludge
21
Fig. 2. SEM observation of dewatered alum sludge
22
7
6
Ce/q (g-sludge/l)
R2 = 0.9985
pH=4.3
pH=6.0
pH=7.0
5
pH=8.5
R2 = 0.9978
pH=9.0
4
3
R2 = 0.9979
2
R2 = 0.9898
R2 = 0.9961
1
0
0
1
2
3
4
5
Ce (mg-P/l)
Fig. 3. Langmuir isotherm plots to determine the maximum adsorption capacities
23
120
adsorption capacity
5
80
Zeta potential
4
40
3
0
2
-40
1
-80
0
-120
4.3
6.0
7.0
8.5
Zeta potential (mV) .
Adsorption capacity .
(mg-P/g-sludge)
6
9.0
pH
Fig. 4. Effect of pH on adsorption capacity and surface charge of alum sludge
24
Adsorption capacity.
(mg-P/g-sludge)
300 mg/l
3
400 mg/l
500 mg/l
2
750 mg/l
1000 mg/l
.
1
0
4.3
6.0
7.0
8.5
9.0
pH
Fig. 5. Effects of Cl- and SO42- on phosphate adsorption capacity at different pH
25
0.8
0.7
Absorbation
0.6
0.5
0.4
0.3
0.2
Before adsorbation
After adsorbation
1400
1200
1000
800
Wavenumber (cm-1)
600
400
Fig. 6. The FT-IR spectra of the sludge samples before and after adsorption
26
Phosphate (mg-P/l).
(a)
8
Adsorption stage
Phosphate addition
(5 mg-P/l)
6
4
Phosphate
2
0
Hydrolysis stage
pH
8
7
6
pH
(d)
(e)
Total aluminium (ug/l) . TOC (mg/l) .
(c)
Cl- and SO4 2- (mg/l) .
(b)
5
8
6
4
Cl
SO
2
0
10
8
6
4
2
0
TOC
30
20
10
Total aluminium
0
0
7
14
21
0
Hydrolysis time (day)
27
2
4
Adsorption time (hour)
6
Fig. 7. Variation of pH, Cl-, SO42-, TOC and total aluminium during the hydrolysis and
P-adsorption process
28